The overarching goal motivating my research has been to characterize biological heterogeneity in cancer and its impact on cancer epidemiology. As an epidemiology doctoral student at the University of Washington (UW), I developed a research portfolio characterizing risk factors for triple-negative breast cancer - a poor prognosis poorly understood breast cancer subtype. Following my graduation from UW in 2010, my research transitioned from the study of breast cancer to the study of colorectal cancer (CRC), and from the study of cancer risk to the study of cancer survival. In extending my research to the study of CRC survival, the goal motivating my work has remained the same but my specific research interests have evolved to encompass areas in which I have not previously been involved. In particular, I have developed an interest in clinical epidemiology and in studies based in the clinical trial setting. The research and training activities proposed for this career development award will allow me to gain the skills, experience, and collaborations necessary to launch an independent academic career in cancer survival, and to transition my research into a more translational realm. Unifying my predoctoral, postdoctoral, and proposed research is an interest in the relationship between cancer biology and cancer epidemiology. Like breast cancer, CRC is a heterogeneous disease that can be classified into biologically distinct subtypes. Unlike breast cancer, however, molecularly-defined subtypes of CRC have not yet been widely characterized. Molecular classifications for CRC subtypes were recently proposed, using information on four tumor markers: microsatellite instability (MSI), CpG island methylation (CIMP), mutations in BRAF, and mutations in KRAS. Different combinations of these four markers are thought to reflect distinct pathways of CRC development. Biological distinctions between CRC subtypes defined by these markers also likely translate to differences in prognosis and prognostic factors, although such differences have not yet been well described. The research objectives of this proposal are to assess the prognostic significance of CRC molecular subtypes defined by joint MSI / CIMP / BRAF / KRAS status, and to assess the impact of modifiable lifestyle factors on clinical outcomes after CRC diagnosis for patients with these subtypes of disease. The ultimate goal of this research is to identify factors that will have clinical relevance in informing CRC survival. In pursuit of these research objectives, I will leverage data from two very rich but very different studies with which I have existing collaborations: the Seattle Colon Cancer Family Registry (S-CCFR) and a phase III randomized clinical trial of stage III colon cancer (N0147). Both studies are well annotated with epidemiologic data and information on MSI, CIMP, and BRAF and KRAS mutation status in CRC patients. Through the proposed research activities, I will use data from both studies to characterize differences in modifiable lifestyle factors (Aim 1a) and survival (Aim 1b) across four CRC subtypes defined by joint MSI / CIMP / BRAF / KRAS status. I will also use these data to assess the relationship between several lifestyle factors (e.g., smoking, alcohol consumption, physical activity) and survival in patients with these four subtypes of CRC (Aims 2-3). Conducting parallel analyses in an observational study (S-CCFR) and in a clinical trial (N0147) will provide opportunities for comparing and contrasting results across study settings with complementary strengths and limitations. This approach of parallel and complementary analyses will also provide me with experience in working with clinical outcomes data collected in different study settings and in working with clinical trials. In conducting the proposed research, will benefit from the rich resources and collaborative environment of the Fred Hutchinson Cancer Research Center (FHCRC). As one of the world's leading cancer research centers, the FHCRC is home to a large number of distinguished epidemiologists, clinical researchers, and laboratory scientists who can provide directed guidance. In particular, I will benefit from the mentorship of Dr. Polly Newcomb (primary mentor), who has considerable expertise in CRC epidemiology and observational studies of cancer survival and who is PI of the S-CCFR. I will also be mentored by Dr. Noel Weiss (co-mentor), who is a seasoned expert in epidemiologic methods and clinical epidemiology. For further guidance, I have enlisted the expertise of an external advisory committee, comprised of specialists in the conduct of oncology clinical trials (Dr. Steven Alberts, who is also the N0147 Study Chair), in the conduct of correlative science across study settings (Dr. Paul Limburg), in the epidemiology and treatment of CRC (Dr. Andrew Chan), and in CRC pathology (Dr. Christophe Rosty). To complement my research and further my career development, I will also work with Dr. Weiss to develop and implement materials for teaching a graduate-level epidemiologic methods course series. The hands-on experience and expertise gained through this project will be supplemented by formal coursework in outcomes research, pharmacoepidemiology, and clinical trial methodologies, and by regular research seminars offered through the FHCRC, the UW, and other programs. Both the FHCRC and the UW place particular emphasis on providing early career scientists with career development opportunities, offering formal mechanisms for mentoring, regular career development seminars, journal clubs, curriculum evaluation, and other networking resources. Thus, the environment of the FHCRC, and the closely affiliated UW, provides an excellent setting in which to launch my independent academic research career.